New 3D Slicer Workshop Announcement and an Engineering Application!
Hello all! The next 3D Slicer workshop is scheduled to run across two afternoons on the 24th and 25th of November. Sign up to participate for free here!
Check out my previous blog posts here and here to see how 3D Slicer can be used to generate 3D models of anatomical features from medical scan data. The recent 3D Printing Showcase held at Melbourne University had some fantastic 3D printed anatomy displays, hopefully some of you were able to come along and gain some inspiration!
I thought I’d supplement this blog post with another example of how 3D Slicer may be used, not in the medical field however, but in the engineering field.
Although 3D Slicer is targeted towards the processing and visualisation of medical scan data, it’s applications aren’t necessarily limited to medical applications. At its core, 3D Slicer is all about image processing, regardless of where those images came from, and although much of the image data 3D Slicer will process comes from DICOM datasets, other sources can be imported just as readily.
I was given an image stack consisting of 99 2 dimensional images (in .tiff file format), of an alumina foam micro-structure. This dataset was created using micro-computed tomography (MCT) and can be used to generate a 3 dimensional representation of the alumina foam structure.
Image: The alumina foam samples, and micro-structure as captured by micro-computed tomography (MCT).
When importing these images into 3D Slicer, they are combined into a volume, and I was immediately able to view the dataset not just in the direction that the images were captured in (z axis) but along the x and y axis. A 3D volume rendering of the composite could also be readily generated.
Image: The alumina foam viewed in 3 perpendicular directions, and a 3D volume rendering of the dataset.
3D Slicer comes with hundreds of basic and advanced image filters from ITK. The Median Image Filter reduces noise in an image while preserving image edges, by replacing the intensity of each pixel with the median intensity of surrounding pixels. This makes subsequent image segmentation (labelling different sample features with different coloured labelmaps) easier. The below image series shows the sample image being filtered, then segmented into two regions, representing the alumina (green) and empty pores (red). A fairly good binary representation of the micro-structure is quickly achieved.
Image: The alumina foam micro-structure filtered, then segmented into two labelmaps.
These labelmaps can then be used to generate a 3D surface model of the two regions, as shown below, which may be visualised virtually, or 3D printed into a physical object.
Image: 3D surface models of the alumina foam, representing the alumina (green) and pores (red).
So, in essence, even if you aren’t in the medical field, you may find 3D Slicer useful for an entirely different purpose!
For more information, feel free to contact me at louisevanderwerff@gmail.com, or tweet me @LouWerff.
